ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017

ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.

Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

AGRICULTURAL SUPPLY CHAIN MANGO INVENTORY MODEL

MODELO DE INVENTARIOS PARA LA CADENA AGRÍCOLA DEL MANGO♣

Javier Arturo Orjuela-Castro1, , Miguel Angel Cardona-Rojas2, Laura Vanessa



Castañeda-López3

ABSTRACT

The globalization has generated changes in world food consumption, demand for healthy
foods as fresh fruits and vegetables have increased. The losses in the perishable fruit supply
chains can reach up to 40%, the post-harvest loss in Co-lombia is mainly due to the inadequate
planning and execution of the inventory processes. The problem is due to the high perishability
of the fruit and its variabil-ity in terms of quality levels, which even under optimal conditions of
storage tend to decrease. After reviewing publications in the field of inventory management
of perishable goods, there is no research regarding fruit supply chain. Additionally, some
models do not cover the entire supply chain, and they only focus in studying one or two actors
involved in the chain. This article presents a management mod-el of multi-level inventory for
fruit mango chain includes functions decay and own losses of highly perishable fruit, a model
nonlinear programming, formulated and evaluated in GAMS, which minimizes the total cost
of inventory in the chain formed by a farmer, an agro-industrial company, a wholesaler, a
hypermarket, a reseller and a retailer allowing calculate the order quantity and time of optimal
re-placement of fruit for each echelon.

Palabras Clave: Inventory Supply Chain, Perishable Fruit, Mango, PNL, Multi-echelon

INTRODUCTION

Controlling inventory is one of the most important logistics operations in agricultural supply
chains (Coelho & Laporte, 2014; Qin et al., 2014; Chung et al., 2014; Duan & Liao, 2013; Salin,
1998) especially for products such fruits and vegetables, due to the high rate of perishability
of these kinds of goods. In fact, fresh fruits even decrease their quality when they are in
motion, and also, fruits offer is seasonal which leads to a volatile demand (Soto-Silva et al.,
2016), permitting consumer choice to be influenced by the availability of the fruit at the time of
purchasing and the quality thereof, (Yang & Wee, 2003; Akkerman et al., 2010; Van Der Vorst,
2006; Duan & Liao, 2013). Therefore, there are great challenges for obtaining an efficient
food supply chain of perishable goods (Batero & Orjuela 2018). Although this has been in the
interest of research regarding inventory management in these supply chains, there are not
models considering perishable food supply chains, particularly not for fruits.

♣The version originally was submitted at IX Simposio Internacional de Ingeniería Industrial: Actualidad y Nuevas Tendencias October 2016- Brasil and improved for this journal after the peer-review process.

1 Department of Industrial Engineering, Universidad Distrital, Bogotá-Colombia, ORCID: http://orcid.org/0000-0003-2187-06302 UFRGS. Porto Alegre. Brazil. orcid.org/0000-0002-5795-4468 ribeiro@

producao.ufrgs.br
2 Faculty of Engineering, Universidad Distrital, Bogotá-Colombia. orcid.org/0000-0003-3061-0021
3 Faculty of Engineering, Universidad Distrital, Bogotá-Colombia. orcid.org/0000-0002-5083-4674

♦Corresponding author: jorjuela@udistrital.edu.co

Received: 22.10.2016 Accepted: 18.04.2017

141
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

Food supply chains are different from other supply chains (Zanoni & Zavanella, 2012),
Starting from the harvest time until the delivery to the final consumer, there is variation in the
quality of the food, which is susceptible to change even under optimum distribution conditions
throughout the entire chain (Sloof et al., 1996; Orjuela et al., 2017), Using integrated planning
models in food supply chain is still a subject very limited in terms of studies, especially
regarding planning models for fresh fruits supply chains, as existing models in the literature
do not incorporate realistic features such lifespan of the product (Manzini & Accorsi, 2013).

Fruits, vegetables, and flowers supply chains may comprise farmers, wholesalers, importers
and exporters, retailers and specialty stores, as well as suppliers of goods and services
(Van Der Vorst et al., 2007). A close relationship between supply and demand is the key for
increasing the competitiveness of such chains (Duan & Liao, 2013). Food supply chains are
global complex networks, covering production, processing, distribution, and even elimination
of damaged food (Yu & Nagurney, 2013) in which logistics operations become important for
agribusiness management, since all of these types of goods are perishable and its short life
cycle makes inventory management a critical aspect (Orjuela-Castro et al., 2017).

Inventory management in the supply chain of agricultural goods is not a topic widely studied,
however, there are some related studies such RAI et al., (2013) who defined an optimal
replenishment time for the retailer in the supply chain, while Pauls-Worm et al., (2013)
studied a farmer facing a non-stationary demand with constrained service level, formulating
replenishment policies for waste inventory. Law & Wee (2006) studied the transformation of
chickens, pigs, and fishes in the supply chain considering a rate of deterioration with a Weibull
distribution, as well as discount policies, late deliveries and multiple orders.

Other studies focus on determining planting time, crop size and also the area to be planted,
taking into account a stochastic demand in order to reflect a more realistic behavior Lodree
& Uzochukwu (2008) consider seasonal sales and the selection of the inventory by the
consumer; while Tan & Çömden (2012) intended to match the supply and demand of a fresh
fruit-vegetable supply chain; additionally, they compare the system’s performance with both,
one and multi-suppliers.

After reviewing publications in the field of inventory management of perishable goods, one can
say there is a small number of articles related to inventory management models for food supply
chain and particularly there is no research regarding fruit supply chain. The complexity of the
problem is due to the high perishability of the fruit and its variability in terms of quality levels,
that even under optimal conditions of storage and distribution tend to decrease. Additionally,
some models do not cover the entire supply chain, and they only focus in studying one or two
actors involved in the chain.

As result of reviewing the state of the art, we used as a base for our research the works
done by (Rau et al., 2003) and (Wang et al., 2011) who considered the behavior of inventory
management for perishables. In this research those studies were supplemented by particular
variables in the formulation of a model, for inventory management in the Mango supply chain
in Cundinamarca-Bogotá, also we included some constraints not contemplated before, as
well as the model covers six echelons of the fruit supply chain, not done before. The model
permits to calculate the losses due to the time in inventories elapsed of the different actors, to
be accumulated and compared with the life cycle of the fruit, after of harvest.

METHODOLOGY
Before the to design the model, we make a diagnostic about the behavior of the management

142
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

of the inventory by the actors of the mango supply chain in Cundinamarca-Bogotá, Colombia,
we use the methodology proposed by (Orjuela-Castro et al., 2016) To the diagnostics were
done a survey, by means of non-probabilistic snowball sampling, 24 producers, 20 wholesalers,
42 retailers (marketplaces and shopkeepers), 10 agroindustry and 3 supermarkets were
surveyed. In order to include the effect of the seasonality of the fruit harvest, we propose a
push system, since production capacity throughout the year is limited and is conditioned to
the maximum crop yields of the farmer in the harvest season. This model is characterized by a
restricted production rate, which depends on prices and supply-demand. The model calculates
both, the cost for each actor independently and the total cost generated in the supply chain.
The deterioration rate of the mango was defined based on the estimated average length of
time in each link of the chain; the respiration rate in the post-harvest maturing stage was
defined according to the stage of life and the properties of the fruit (Figure 1). We estimated
the time taken for each actor in the chain, the maximum and minimum lifetime of permanence,
which was iterated until reaching a value that converges to the average rate of deterioration
of the fruit.

Figure 1. Mango Respiration (Orjuela-Castro et al., 2017).

For the formulation of the models, we established the following assumptions:

One single product is sold through the chain. One replenishment cycle is considered.
Replacement of damaged product is not allowed. Replacement rate is instantaneous. The
Shortage is not allowed. Demand rate is deterministic. The Planning horizon is determined
by the lifetime of the fruit. Ability to receive and send products is unlimited. Production rate is
finite and greater than the sum of all demands from buyers. Each actor in the supply chain
must meet a minimum demand per cycle.

The rate of deterioration of the product depends on the actor, it’s constant and is also known.
This was calculated from the study of the respiration rate of the fruit (Orjuela-Castro et al.,
2017).

The following notations are used for the decision variables and the parameters considered in
the model.

Notation for the variables used in the model

For the farmer (a)

ta : Production cycle time.
Qa: Harvest size of the farmer during t_a

143
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

For the wholesaler (m)

tm : Replacement cycle time for the wholesaler.
Qm: Harvest size of the farmer during t_a

For the agribusiness (g)

tg : Replacement cycle time for agribusiness.
Qg: Order size for fresh fruit during t_g for the agrobusiness

For the hypermarket (h)

th : Replacement cycle time for hypermarket.
Qh: Order size for fresh fruit during t_h for the hypermarket

For the retailer (r)

tr : Replacement cycle time for the retailer days (d).
Qr: Order size for fresh fruit during t_r for the retailer (Kg)

For the reseller (v)

tv : Replacement cycle time for the reseller.
Qv: Order size for fresh fruit during t_v for the reseller

Fruit chain
CT: Total logistics costs generated in the fruit chain

The notation for parameters used in the model can found on the annex one

Inventory model formulation

Figure 2 exhibits the different relationships in terms of generation of orders between the actors
in the supply chain, which is composed by the Farmer (Q1) Wholesaler (Q2), Agribusiness
(Q3), Hypermarket (Q4), Retailer (Q5), and Resellers (Q6).

Figure 2. Composition of the agoindustry fruit supply chain

f The total cost of the chain is represented by equation (1), which corresponds to the total sum

144
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

of the logistics costs generated for each actor in the chain. Equation (2) describes the number
of optimal orders that each actor must order to its supplier(s), likewise, equation (3) ensures
that the mango order size is greater or equal to the demand exhibited, so that demand can
be satisfied.
6
MinimizeTC = ∑ CTi (ti ) (1)
i =1

Constraints:
(2)
Di
=Qi *(eθiti − 1)
θi
(3)
Qi ≥ Di

Equations (4) and (5) represent the total cost for the farmer and other actors respectively,
corresponding to the sum of the cost of maintaining inventory, deterioration, ordering and
losses arising from the imbalance between supply and demand (in the case of the farmer the
cost of producing and harvesting is added, and there is no cost of carrying out orders).

(ta ) CM a * Pa + (CDa * Pa *θ a ) + Ca + Cu * Qa + La * PV
CTa= (4)

(Qi − ( Di * ti ) (5)
CTi (ti ) CM i *
= + CDi *(Qi − ( Di * ti )) + Coi + Li * PV
θi

Equations (6) and (7) represent the demands to be met by the farmer and the wholesaler for
each of the actors they supply to.
(6)
D1 = 1.22* ∑ 2 Qi1
4

D2 = 1.22* ∑ 3 Qi 2
6 (7)

Equation (8) represents the maximum time that the fruit can remain in each link of the chain;
in order not to exceed the time within the fruit can be consumed. Meanwhile, equation (9)
considers that this time must be greater than zero. Where X = 1 when there is flow of fruit and
X = 0 when there is not flow.
(8)

T ≥ ∑ 1 ti * X i
6

(9)
ti ≥ 0

Equation (10) shows the losses in each link of the chain, due to the gaps presented between
supply and demand.

∑ ∑
2 6 (10)
=Pi 1 2
Qi − Q j

RESULTS - CASE STUDY

Mangos are climacteric fruits exhibiting seasonality in their harvest. This aspect is highly
considered when making relevant decisions for inventory management within the actors of
the chain (Figure 2).

From surveys conducted during 2015-2016, specifically for the case of Cundinamarca-Bogotá,
we set the configuration of the supply chain. Given that mango production is seasonal, the

145
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

agro-industrial supply chain was modeled as a push system. Table 1 shows the results after
applying the optimization model. This table exhibits the average inventory to be maintained
by each actor in the chain. The wholesaler has the highest storage needs, either in transit or
in storage.

Table 1. Tipos de capacidades dinâmicas.

Actor Ii (kg)
Farmer (a)
Wholesaler (m) 13,44
Agribusiness (g) 4,76
Hypermarket (h) 6,07
Retailer (r) 4,79
Reseller (v) 5,65

Table 2 shows the main optimal values obtained. There is high production by the farmer
because it’s harvest time, so the size of the crop is 54,23 kg mango, with a cost of US$10,8
On the other hand, it is evident that the retailer has the minimum time replen-ishment of
inventories because it’s ordering is a small lot size.

Table 2. Optimal results

Actor Qi*(kg) ti*(d) CTi*(US$) Di(kg) CT*(US$)

Farmer (a) 54,23 - 10,81 -
Wholesaler (m) 27,34 1 4,15 26
19,38
Agribusiness (g) 8,13 1,16 0,75
Hypermarket (h) 13,33 0,93 1,51 13,33
Retailer (r) 10,71 0,92 1,07 10,71
Reseller (v) 12,5 0,93 1,09 12,50

Table 3 shows the losses caused by the imbalance between supply and demand, given that
the farmer sends a harvest size to the actors downstream the chain hoping that all the fruit is
received, however the order size of each actor is different from the expected, therefore, there
are declines in shipping, which are supported by the farmer and the wholesaler.

Table 3. Losses associated with the imbalance between supply and demand.

Chain Parameter Losses

(Kg)
7,5% 5,52
Farmer 5% 5,42
(a)
5,33
4,84

146
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

Wholesaler 4,13
(m) 5% 3,45

Consequently, with a rate of decline of 5% there will be a loss of 5,42 and 4,13 kg caused by
lack of demand, in the case of the farmer and the wholesaler respectively.

Sensitivity Analysis
Table 4 shows a change in the size of optimal ordering when the rate of deterioration of the
mango changes.

Table 4. Effects of the deterioration rate over the cost in the mango supply chain

Qi*(kg) ti*(d) CTi*(US$) Ii (kg) Di(kg/d) CT* (US$)

θa= 7,5% 55,17 11,10 21,12
(a) θa= 5% 56,36 10,81 19,38
θa= 2,5% 53,31 10,54 17,62
θm= 15% 28,05 1 4,99 13,67 26 21,12
(m) θm=10% 27,34 1 4,15 13,44 26 19,38
θm= 5% 26,66 1 3,35 13,22 26 17,62
θg= 12% 8,27 1,16 0,,84 4,68 21,12
(g) θg= 8% 8,13 1,16 0,75 4,76 19,38
θg= 4% 8,00 1,17 0,66 4,65 17,62
θh= 22,5 13,33 0,90 1,65 5,81 13,3 21,12
(h) θh= 15% 13,33 0,93 1,51 6,07 13,3 19,38
θh= 7,5% 13,33 0,96 1,36 6,35 13,3 17,62
θr= 27% 10,71 0,88 1,27 4,55 10,7 21,12
(r) θr= 18% 10,71 0,92 1,07 4,79 10,7 19,38
θr= 9% 10,71 0,96 0,85 5,05 10,7 17,62
θv= 24% 12,5 0,90 1,29 5,40 12,5 21,12
(v) θv= 16% 12,5 0,93 1,09 5,65 12,5 19,38
θv= 8% 12,5 0,96 0,86 5,93 12,5 17.62

When the rate of decline decreases, the lot size must be less for the farmer and wholesaler,
due to losses by deterioration. Also, it is evident that the lower the rate of deterioration the
higher the time for replacement of the mango, leading to a longer period of the quality of the
mango, which increases the average inventory of every actor.

From all the analyses performed, lower rates of deterioration showed a better performance in
terms of costs, cycle times and replenishment quantities, suggesting that with good agricultural
practices, cold chains and modified atmospheres may decrease the rate of respiration of
fruit, improving performance measures, thus benefiting all the fruit supply chain and reducing
inventory costs for every actor.

147
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

CONCLUSIONS
We present a model to management of inventories which aims to minimize the total cost
of inventory in the supply chain, considering losses by deterioration, cost of harvest for the
farmer, cost of ordering in a fruit perishable supply chain, integrates inventory costs for the
farmer, wholesaler, agroindustry, hypermarket, retailer, and reseller all involved within the
mango chain. The model presented was done using a push system, taking into account
crop conditions. Through the proposed model we obtained optimal results of replenishment
times, crop lot size and order size, which allowed improving mango logistics supply chain
performance on Cundinamarca-Bogota, Colombia.

We measured the effect of the rate of deterioration of the mango over the decisions for
ordering. We found that with lower rates of decline, costs tend to reduce, and an improvement
occurs reflected in the cycle times and replenishment quantities, leading to the conclusion
that reducing the respiration rate of the fruit, is possible to obtain efficient agricultural chains,
with lower costs for each actor involved.
The model for the mango chain was the result of reviewing the state of the art, although we
took as pillar studies those developed previously, however, this model contains differences
regarding the rate of deterioration which depends on the fruit, the losses due to inventory time
in the different actors, novel approach.

FUTURE RESEARCH
For future research, it is important to analyse the impact of stochastic elements presented in
the demands for actors such as the hypermarket, retailer and reseller and what its effect on
the price of the mango is.

On the other hand, there is evidence from the literature review, that there is the need to develop
models that consider more than a logistic process simultaneously, such as location, transfer,
routing and inventory and its application in the Colombian context, since replenishment lead
times were not taken into account and this is a determining factor because of the conditions
of the road infrastructure and fruit collection processes as well for the selection centers owned
by the region.

Finally, the development of a dynamic model that considers the number of orders in a finite
planning horizon is suggested. On the other hand, one might consider the classical models
to model the supply chain even when the products lose their quality and become perishable.

REFERENCES
AKKERMAN, R., FARAHANI, P., and GRUNOW, M. Quality, safety and sustainability in food
distribution: a review of quantitative operations management approaches and challenges. Or
Spectrum, 2010, 32(4), 863-904.

BATERO-MANSO, Diego Fernando and ORJUELA-CASTRO, Javier Arturo. Inventory

Routing Problem in perishable supply chains: A review of the state of the art. Ingeniería,
2018, vol. 23, no 2

148
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

CHUNG, K.J., CÁRDENAS-BARRÓN, L.E. and TING, P.S. An inventory model with non-
instantaneous receipt and exponentially deteriorating items for an integrated three layer
supply chain system under two levels of trade credit. International Journal of Production
Economics , 2014, 155, 310-317.

COELHO, L. and LAPORTE, G. Optimal joint replenishment, delivery and inventory

management policies for perishable products. Computers & Operations Research, 2014, 47,
42-52.

DUAN, Q. and LIAO, T.W. A new age-based replenishment policy for supply chain inventory
optimization of highly perishable products. International journal of production economics ,
2013, 145(2), 658-671.

LAW, S.T. and WEE, H.M. An integrated production-inventory model for ameliorating and
deteriorating items taking account of time discounting. Mathematical and Computer Modelling,
2006, 43(5), 673-685.

LODREE, E. J., and UZOCHUKWU, B. M.. Production planning for a deteriorating item with
stochastic demand and consumer choice. nternational Journal of Production Economics,
2008,116(2), 219-232.

MANZINI, R. and ACCORSI, R. The new conceptual framework for food supply chain
assessment. Journal of Food Engineering, 2013, 115(2), 251-263.

ORJUELA-CASTRO, J.A., DIAZ-RIOS, O.J. GONZÁLEZ PEREZ, A. Logistics’ diagnostic in

cosmetics and toiletries supply chain. Revista Cientifica, 2016, 1(28), 81-96.

ORJUELA-CASTRO, J.A., DIAZ-GAMEZ, G.L. and BERNAL-CELEMÍN, M.P. Model for

Logistics Capacity in the Perishable Food Supply Chain. Communications in Computer and
Information Science, In Workshop on Engineering Applications. Springer, Cham, 2017, 742,
225-237.

ORJUELA-CASTRO, J.A. ADARME-JAIMES, W. Dynamic impact of the structure of the

supply chain of perishable foods on logistics performance and food security. Journal of
Industrial Engineering and Management, 2017, 10(4), 687-710.

ORJUELA-CASTRO, Javier Arturo; MORALES-AGUILAR, Fredy Santiago; MEJÍA-FLÓREZ,

Laura Fernanda. Which is the best supply chain for perishable fruits, Lean or Agile?. Revista
Colombiana de Ciencias Hortícolas, 2017, vol. 11, no 2, p. 294-305

PAULS-WORM, K.G., HENDRIX, E., HAIJEMA, R. and VAN DER, G. Inventory control for a
perishable product with non-stationary demand and service level constraints. 2013. Obtenido
de optimization-online: Online, www. optimization-online. org/DB FILE/2013/08/4010

QIN, Y., WANG, J. and WEI , C. Joint pricing and inventory control for fresh produce and
foods with quality and physical quantity deteriorating simultaneously. International Journal of
Production Economics, 2014, 152, 42-48.

RAI, S. K., SINGH, V. and VAISH, A. Persishable food inventory management: A retailer’s
perspective. 2013, 2(2), 1–10.

RAU, H., WU, M.Y. and WEE, H.M. Integrated inventory model for deteriorating items under a

149
ISSN 0717-9103 revista Ingeniería Industrial-Año 16 Nº2: 141-150, 2017
ISSN Online 0718-8307 Agricultural Supply Chain ….Orjuela-Castro et al.
Universidad del Bío-Bío https://doi.org/10.22320/S07179103/2017.09

multi-echelon supply chain environment. International journal of production economics, 2003,

86(2), 155-168.

SALIN, V. Information technology in agri-food supply chains. The International Food and
Agribusiness Management Review, 1998, 1(3), 329-334.

SLOOF, M., TIJSKENS L, M.M. and WILKINSON, E.C. Concepts for modelling the quality of
perishable products. Trends in Food Science & Technology, 1996, 7(5), 165-171.

SOTO-SILVA, W.E., NADAL-ROIG, E., GONZÁLEZ-ARAYA, M.C. and PLA-ARAGONES, L.

M. Operational research models applied to the fresh fruit supply chain. European Journal of
Operational Research , 2016, 251(2), 345-355.

TAN, B. and ÇÖMDEN, N. Agricultural planning of annual plants under demand, maturation,
harvest, and yield risk. European Journal of Operational Research, 2012, 220(2), 539-549.

VAN DER VORST, J.G., DA SILVA, C. and TRIENEKENS, J. H. Agro-industrial supply chain
management: concepts and applications. 2007. FAO.

VAN DER VORST, J. Performance measurement in agri-food supply-chain networks .

Quantifying the agri-food supply chain, 2006, 15-26.

WANG, K.J., LIN, Y. S. and JONAS, C.P.. Optimizing inventory policy for products with
time-sensitive deteriorating rates in a multi-echelon supply chain. International Journal of
Production Economics, 2011, 130(1), 66-76.

YANG, P.C. and WEE, H.M. An integrated multi-lot-size production inventory model for
deteriorating item. Computers & Operations Research, 2003, 30(5), 671-682.

YU, M. and NAGURNEY, A. Competitive food supply chain networks with application to fresh
produce. European Journal of Operational Research , 2013, 224(2), 273-282.

ZANONI, S. and ZAVANELLA, L. Chilled or frozen? Decision strategies for sustainable food
supply chains. International Journal of Production Economics, 2012, 140(2), 731-736.

150